Bartlett Spectrum Research Articles

Spectral Domain Sparse Representation for DOA Estimation of Signals with Large Dynamic Range.

Recently, we proposed a Spectral Domain Sparse Representation (SDSR) approach for the direction-of-arrival estimation of signals incident to an antenna array. In the approach, sparse representation is applied to the conventional Bartlett spectra obtained from snapshots of the signals received by the antenna array to increase the direction-of-arrival (DOA) estimation resolution and accuracy. The conventional Bartlett spectra has limited dynamic range, meaning that one may not be able to identify the presence of weak signals in the presence of strong signals. This is because, in the conventional Bartlett spectra, uniform weighting (window) is applied to signals received by various antenna elements. Apodization can be used in the generation of Bartlett spectra to increase the dynamic range of the spectra. In Apodization, more than one window function is used to generate different portions of the spectra. In this paper, we extend the SDSR approach to include Bartlett spectra obtained with Apodization and to evaluate the performance of the extended SDSR approach. We compare its performance with a two-step SDSR approach and with an approach where Bartlett spectra is obtained using a low sidelobe window function. We show that an Apodization Bartlett-based SDSR approach leads to better performance with just single-step processing.

Application of Sparse Representation to Bartlett Spectra for Improved Direction of Arrival Estimation.

A new technique for high-resolution direction of arrival estimation is presented. The method utilizes the traditional Bartlett spectra and sparse representation to locate emitters in single and multiple emitter scenarios. A method for selecting the sparse representation regularization parameter is also presented. Using Monte Carlo simulations, we show that the proposed approach achieves accurate direction of arrival (DOA) estimations that are unbiased and a variance that approaches the Cramer–Rao lower bound. We show that our method outperforms the popular MUSIC algorithm, and is slightly better than the sparse representation based L1-SVD algorithm when angular separation between emitters is small, signal SNR is low, and a small number of snapshots are used in DOA estimation.

Coincidence functions and Bartlett spectra of point processes

The second order statistical properties of time point processes (PP) are described by the time coincidence function (CF) and the frequency Bartlett spectrum (BS). For PPs recorded by pulses appearing at random time instants, as in photodetection experiments, the CF can be measured by various physical devices showing in particular the famous bunching effect of photons. On the other hand, for PPs recorded by the intervals between successive points (lifetimes), especially for renewal processes, there is no usual procedure for the estimation of the CF, and the aim of this paper is to describe an approach of this problem. The starting point is a mathematical relation between the CF and the set of probabilities density functions of the lifetimes of any order of the PP. As a consequence, the CF can be obtained by processing the results of the multiple normalized histograms of these lifetimes. In the cases, relatively rare, where the mathematical expression of the CF is known in closed form, the correct behavior of the procedure is verified by an experimental analysis of simulated data. The method is extended in order to verify the relationship between the CF and the BS.

Time-frequency analysis of locally stationary Hawkes processes

Locally stationary Hawkes processes have been introduced in order to generalise classical Hawkes processes away from stationarity by allowing for a time-varying second-order structure. This class of self-exciting point processes has recently attracted a lot of interest in applications in the life sciences (seismology, genomics, neuro-science, …), but also in the modeling of high-frequency financial data. In this contribution, we provide a fully developed nonparametric estimation theory of both local mean density and local Bartlett spectra of a locally stationary Hawkes process. In particular, we apply our kernel estimation of the spectrum localised both in time and frequency to two data sets of transaction times revealing pertinent features in the data that had not been made visible by classical non-localised approaches based on models with constant fertility functions over time.

Smoothed Spectra, Ogives, and Error Estimates for Atmospheric Turbulence Data

A systematic evaluation is conducted of the smoothed spectrum, which is a spectral estimate obtained by averaging over a window of contiguous frequencies. The technique is extended to the ogive, as well as to the cross-spectrum. It is shown that, combined with existing variance estimates for the periodogram, the variance—and therefore the random error—associated with these estimates can be calculated in a straightforward way. The smoothed spectra and ogives are biased estimates; with simple power-law analytical models, correction procedures are devised, as well as a global constraint that enforces Parseval’s identity. Several new results are thus obtained: (1) The analytical variance estimates compare well with the sample variance calculated for the Bartlett spectrum and the variance of the inertial subrange of the cospectrum is shown to be relatively much larger than that of the spectrum. (2) Ogives and spectra estimates with reduced bias are calculated. (3) The bias of the smoothed spectrum and ogive is shown to be negligible at the higher frequencies. (4) The ogives and spectra thus calculated have better frequency resolution than the Bartlett spectrum, with (5) gradually increasing variance and relative error towards the low frequencies. (6) Power-law identification and extraction of the rate of dissipation of turbulence kinetic energy are possible directly from the ogive. (7) The smoothed cross-spectrum is a valid inner product and therefore an acceptable candidate for coherence and spectral correlation coefficient estimation by means of the Cauchy–Schwarz inequality. The quadrature, phase function, coherence function and spectral correlation function obtained from the smoothed spectral estimates compare well with the classical ones derived from the Bartlett spectrum.

On long-range dependence of random measures

AbstractThis paper deals with long-range dependence of random measures on ℝd. By examples, it is demonstrated that one must be careful in order to define it consistently. Therefore, we define long-range dependence by a rather specific second-order condition and provide an equivalent formulation involving the asymptotic behaviour of the Bartlett spectrum near the origin. Then it is shown that the defining condition may be formulated less strictly when the additional isotropy assumption holds. Finally, we present an example of a long-range dependent random measure based on the 0-level excursion set of a Gaussian random field for which the corresponding spectral density and its asymptotics are explicitly derived.

Locally stationary Hawkes processes

This paper addresses the generalization of stationary Hawkes processes in order to allow for a time-evolving second-order analysis. Motivated by the concept of locally stationary autoregressive processes, we apply however inherently different techniques to describe the time-varying dynamics of self-exciting point processes. In particular we derive a stationary approximation of the Laplace functional of a locally stationary Hawkes process. This allows us to define a local mean density function and a local Bartlett spectrum which can be used to compute approximations of first and second order moments of the process. We complete the paper by some insightful simulation studies.

Cumulants of Hawkes point processes.

We derive explicit, closed-form expressions for the cumulant densities of a multivariate, self-exciting Hawkes point process, generalizing a result of Hawkes in his earlier work on the covariance density and Bartlett spectrum of such processes. To do this, we represent the Hawkes process in terms of a Poisson cluster process and show how the cumulant density formulas can be derived by enumerating all possible "family trees," representing complex interactions between point events. We also consider the problem of computing the integrated cumulants, characterizing the average measure of correlated activity between events of different types, and derive the relevant equations.

Second-Order Characteristics of the Edge System of Random Tessellations and the PPI Value of Foams

The mean number of pores per inch (PPI) is widely used as a pore size characteristic for foams. Nevertheless, there is still a lack of fast and reliable methods for estimating this quantity. We propose a method for estimating the PPI value based on the Bartlett spectrum of a dark field image of the material. To this end, second-order properties of the edge systems of random tessellations are investigated in detail. In particular, we study the spectral density of the random length measure of the edges. It turns out that the location of its first local maximum is proportional to the PPI value. To determine the factor of proportionality, several random tessellation models as well as examples of real foams are investigated. To mimic the image acquisition process, 2D sections and projections of 3D tessellations are considered.

CHARACTERIZATION OF THE FORMATION OF FILTER PAPER USING THE BARTLETT SPECTRUM OF THE FIBER STRUCTURE

The formation index of filter paper is one of the most important characteristics used in industrial quality control. Its estimation is often based on subjective comparison chart rating or, even more objective, on the power spectrum of the paper structure observed on a transmission light table. It is shown that paper formation can be modeled as Gaussian random fields with a well defined class of correlation functions, and a formation index can be derived from the density of the Bartlett spectrum estimated from image data. More precisely, the formation index is the mean of the Bessel transform of the correlation taken for wave lengths between 2 and 5 mm.

Bartlett spectrum and mixing properties of infinitely divisible random measures

We prove that the Bartlett spectrum of a stationary, infinitely divisible (ID) random measure determines ergodicity, weak mixing, and mixing. In this context, the Bartlett spectrum plays the same role as the spectral measure of a stationary Gaussian process.

Bartlett spectrum and mixing properties of infinitely divisible random measures

We prove that the Bartlett spectrum of a stationary, infinitely divisible (ID) random measure determines ergodicity, weak mixing, and mixing. In this context, the Bartlett spectrum plays the same role as the spectral measure of a stationary Gaussian process.

The pair correlation function of spatial Hawkes processes

Spatial Hawkes processes can be considered as spatial versions of classical Hawkes processes. We derive the pair correlation function of stationary spatial Hawkes processes and discuss the connection to the Bartlett spectrum and other summary statistics. Particularly, results for Gaussian fertility rates and the extension to spatial Hawkes processes with random fertility rates are discussed.

Power spectra of random spike fields and related processes

In this article, we review known results and present new ones concerning the power spectra of large classes of signals and random fields driven by an underlying point process, such as spatial shot noises (with random impulse response and arbitrary basic stationary point processes described by their Bartlett spectra) and signals or fields sampled at random times or points (where the sampling point process is again quite general). We also obtain the Bartlett spectrum for the general linear Hawkes spatial branching point process (with random fertility rate and general immigrant process described by its Bartlett spectrum). We then obtain the Bochner spectra of general spatial linear birth and death processes. Finally, we address the issues of random sampling and linear reconstruction of a signal from its random samples, reviewing and extending former results.

Power spectra of random spike fields and related processes

In this article, we review known results and present new ones concerning the power spectra of large classes of signals and random fields driven by an underlying point process, such as spatial shot noises (with random impulse response and arbitrary basic stationary point processes described by their Bartlett spectra) and signals or fields sampled at random times or points (where the sampling point process is again quite general). We also obtain the Bartlett spectrum for the general linear Hawkes spatial branching point process (with random fertility rate and general immigrant process described by its Bartlett spectrum). We then obtain the Bochner spectra of general spatial linear birth and death processes. Finally, we address the issues of random sampling and linear reconstruction of a signal from its random samples, reviewing and extending former results.

Spectral theory for random closed sets and estimating the covariance via frequency space

A spectral theory for stationary random closed sets is developed and provided with a sound mathematical basis. The definition and a proof of the existence of the Bartlett spectrum of a stationary random closed set as well as the proof of a Wiener-Khinchin theorem for the power spectrum are used to two ends. First, well-known second-order characteristics like the covariance can be estimated faster than usual via frequency space. Second, the Bartlett spectrum and the power spectrum can be used as second-order characteristics in frequency space. Examples show that in some cases information about the random closed set is easier to obtain from these characteristics in frequency space than from their real-world counterparts.

Spectral theory for random closed sets and estimating the covariance via frequency space

A spectral theory for stationary random closed sets is developed and provided with a sound mathematical basis. The definition and a proof of the existence of the Bartlett spectrum of a stationary random closed set as well as the proof of a Wiener-Khinchin theorem for the power spectrum are used to two ends. First, well-known second-order characteristics like the covariance can be estimated faster than usual via frequency space. Second, the Bartlett spectrum and the power spectrum can be used as second-order characteristics in frequency space. Examples show that in some cases information about the random closed set is easier to obtain from these characteristics in frequency space than from their real-world counterparts.

Properties of the maximum likelihood and pisarenko spectral estimates

The performance of the maximum likelihood spectral estimate is compared vis-à-vis the classical Bartlett spectrum. Asymptotic behavior and special cases such as single tone harmonic analysis are considered. Also, the Pisarenko method of harmonic retrieval is analyzed for the case of a single tone imbedded in noise. The statistical properties of the eigenvector corresponding to the minimum eigenvalue as well as that of the estimated frequency value are derived. The dependence of the reliability of the estimate upon the data size and the tone frequency location is illustrated.